Objective: The purpose of this study was to verify changes in a driver’s emotions through the physical characteristics of haptic signals. This is to improve the performance of drivers by designing haptic signals with emotional semantics. Background: Currently, drivers receive a variety of information through intelligent systems installed in their vehicles. Because this is mainly achieved through visual and auditory channels, an excessive amount of information is provided to drivers, which increases the amount of information and cognitive load that they must accept. This, in turn, can reduce driving safety. It is, therefore, necessary to develop a haptic signal, a sensory channel that has not been widely used in in-vehicle information systems. Methods: The experiment was performed to collect a driver’s emotions according to the haptic signal in a driving simulator. Haptic signals were designed by various frequencies and accelerations, and driver emotions were collected through Kansei engineering techniques and analyzed through factor analysis. To verify intelligibility, haptic signals were compared and evaluated based on response time, response rate, and amount of transmitted information. Results: The final determined emotional map consisted of dangerousness and urgency. Based on the emotional map, four emotional semantic haptic signals were designed. It was confirmed that these four signals displayed higher performance than the discriminability haptic signal in terms of response time, response rate, and amount of transmitted information. Conclusions: Using emotional maps, it is possible to design haptic signals that can be applied to various driving situations. These maps may also assist in securing design guidelines for haptic signals that apply to in-vehicle information systems.